The OpenClaw frenzy that spread throughout China is testing Beijing’s approach to artificial intelligence, as it seeks to balance the push for cutting-edge innovation against concerns over data security.
Tech giants and consumers have rushed to embrace the agentic AI phenomenon in the past few weeks and at its peak hundreds of people queued outside a tech shop in Shenzhen for help installing the software.
But runaway uptake of the foreign technology caught the attention of officials and on Wednesday the central government warned state-run enterprises and agencies against installing it on office computers, according to people familiar with the matter.
China is racing to establish itself as a world leader in AI and has pushed to twitter-tweet it across its economy, with an aim to transform industries as well as everyday life. The new directive underlines how that ambition is jutting up against the ruling Communist Party’s instinct for stability and control.
“Chinese regulators typically respond with extraordinary speed to threats from emerging technologies, but the rate of adoption of OpenClaw and other agentic tools is still outpacing them,” said Kendra Schaefer, partner and director of tech policy research at Trivium China.
Software like OpenClaw, an open-source autonomous AI agent developed by an Austrian which slipped into mainstream use this year, can carry out tasks from email clean-ups to calendar management and flight check‑ins. The practice of installing the AI agent has even acquired a nickname in China: “raising lobsters.”
Tencent, Alibaba, MiniMax, and Baidu have all launched OpenClaw-compatible tools. Local governments in cities including Shenzhen, Wuxi, and Hefei have announced large subsidies for startups building atop the platform. But it requires unusually broad access to private data and can communicate externally, potentially exposing computers to external attack.
Yin Tongyue, the chairman of Chery Automobile Co. Ltd, one of China’s biggest electric vehicle manufacturers, said during the frenzy he asked his staff to hold off from using the software until there was a training program. “I said that everyone should hold off on the installation for now and we can have a focused training program. We should be willing to embrace new things, but not follow the crowd blindly… it can lead to some risks beyond our imagination”, he told Bloomberg News.
Regulatory Response
Beijing has previously raised the alarm about foreign actors targeting datasets including geospatial and genetic information and the speed of OpenClaw adoption is adding to the urgency for regulatory response.
The Chinese government has eschewed a sweeping AI law, unlike the European Union. Instead, since 2022 it has imposed ad-hoc measures to address specific issues, such as rules on algorithm recommendations and deepfake content. Last year, it became the first government to mandate labels for AI-generated content.
“Beijing’s biggest challenge in regulating AI is the same one all governments face: the technology is moving so quickly that a regulation could be out of date before the ink is dry,” said Matt Sheehan, a senior fellow at the Carnegie Endowment for International Peace.
So far there is no regulation specifically targeting this latest software, tackling questions such as who is responsible for the actions of AI agents. On Wednesday, the China Academy of Information and Communications Technology, a research institute affiliated with the Ministry of Industry and Information Technology, said that it plans to trial the trustworthiness of AI agents like OpenClaw starting late March and that it would develop a series of standards for its use.
“I definitely worry about security,” said Ryan Xie, a Chinese teacher in the southern city of Jiangmen who has been using the AI agent to help with repetitive tasks like making presentation slides and filling forms. “That’s why I look for workarounds — like running Openclaw within a Docker container and a Sandbox, or configuring specific rules to restrict it from overstepping its bounds.”
In a sign of the tech sector’s rising importance to the economy, Beijing has set a target of increasing the value added of core digital economy industries to 12.5% of gross domestic product by 2030, from 10.5% last year.
That’s where AI and “lobster fever” presents another challenge: social stability. Automation threatens to displace the world’s largest workforce, in a country already grappling with a fragile labor market and youth unemployment that has hovered above 15% for the past six months.
Lu Jianhua, an academic at the Chinese Academy of Sciences and professor at Tsinghua University, said he has been using AI to research low-altitude economy infrastructure — work that once required a small team of researchers.
“AI serves as a very capable assistant — equivalent to several human assistants,” said Lu, a member of the Chinese People’s Political Consultative Conference.
A recent Peking University study of over a million job postings found that sectors most exposed to AI, including accounting, editing, and programming, are already seeing recruitment declines. In January, the Ministry of Human Resources announced it was drafting policy guidance to address AI’s impact on jobs, though it did not provide a date for its release.
Victor Chen, a fintech worker in Guangzhou who has been using OpenClaw since early February to help with several projects, from developing mobile games to writing novels, also frets about job losses.
“The more significant underlying factor is that the government — like elsewhere in the world — is not actually ready to deal with AI-driven mass unemployment and the social unrest it might cause,” he said.
With assistance from Jing Li, Colum Murphy and Tian Ying.
This article was generated from an automated news agency feed without modifications to text.
